@t_NYC@OutFrontCNN@j_g_allen This is actually connected to the point I'm trying to make. The general comment has been this isn’t as bad as Covid, but that recommendation odds the same one they used last time.
@om_patel5 It looks like he asked Claude how the tokenizer works, and whether it can be fixed not Anthropic. That needs to be fact checked before you make any claims about this. Do you have any evidence from a human to support the claims?
@itsPaulAi Agree on one level, however I also suspect that after a while… There was a piece of Google a ways back that predicted virus case appearance in hospitals more accurately than other methods, and they were only using geographic search term prevalence for source data.
@kvlly Interesting that this article moved to humanizing, with the term "hire," to imply equivalence and a rights / responsibilities model. Hiring the system vs purchasing from the manufacturer? Do we hire an insurance claims system?
Yes, and... There is some value in learning tidbits, but confusing that with capability and expertise is the trip down the slippery slope of unconscious incompetence.
@GaryMarcus How do you propose to solve the problem that not all societies have the same values, and some are fully diametrically opposed to each other in key areas?
@kvlly Maybe the 301M results is the end of the stack they could accommodate before the code faulted staring at itself in the mirror? 😂
That's a great shot!
To improve your writing, read more.
To improve your thinking, write more.
To improve your storytelling, present more.
To improve your energy, rest more.
To improve your understanding, teach more.
To improve your network, give more.
To improve your happiness, appreciate more.